Preoperative Cardiovascular Evaluation for Noncardiac Surgery An Update

1
Preoperative Cardiovascular Evaluation for
Noncardiac Surgery - An Update

• February 27, 2002
• Robert B. Preli

2
Overview

• Scope of the Problem
• Purpose of Clinical Practice Guidelines
• ACC/AHA 2002 Guideline Update/Review
• The ACP Guideline
• Validation Study
• Lees Revised Cardiac Risk Index
• Conclusions

3
Scope of the Problem - Epidemiology

• 27 million patients per year in the United States
  are given anesthesia for surgeries
• 8 million have known CAD or risk factors
• 50,000 who undergo noncardiac surgery will have a
  perioperative MI
• 1 million will have a perioperative complication
• Common reason for consultation during AIM rotation

4
Scope of the ProblemPurpose of Assessment

• Revascularization before noncardiac surgery to
  enable the patient to get through the procedure
  is appropriate for only a small subset of
  patients.
• Determine who can go directly to surgery
• Determine who
• have additional tests prior to surgery
• needs medical management prior to surgery
• Risk stratify patients, NOT clear them for
  surgery

5
Practice Guideline Characteristics

• Comprehensive review of the evidence
• Expert opinion used to make value judgements
• Official endorsement by an organization
• Intention to influence your practice patterns

6
ACC/AHA Guidelines

• Guidelines for Perioperative Cardiovascular
  Evaluation for Noncardiac Surgery
• First published in 1996 by the ACC/AHA
• Updated in January, 2002
• Guidelines are available at
• www.acc.org
• www.americanheart.org

7
Preop Cardiac Assessment
8
Exercise Tolerance
9
Surgical Risk
10
2002 Guideline Highlights

• New intermediate clinical predictor is Renal
  Insufficiency
• Preoperative creatinine gt 2.0 mg/dl
• New Shortcut to Noninvasive Testing table to
  assist in clinical decision making
• New When and Which Test table to assist in
  choosing testing modality
• Strong endorsement for beta blocker use in the
  appropriate patient population
• More than 400 new relevant articles went into the
  2002 guidelines

11
(No Transcript)
12
Revascularization

• In general, limit to patients who have a clearly
  defined need independent of surgery
• Preoperative CABG
• Should be reserved for high-risk patients
  undergoing high risk procedures in whom long-term
  outcome would be improved by CABG
• Preoperative PCI
• Indications are identical to the general ACC/AHA
  guidelines for the use of PTCA
• Wait 2-4 weeks after stent placement for elective
  noncardiac surgery

13
Revascularization

• Coronary Artery Surgery Study (CASS)
• 24,959 patients in database from the NHLBI
  between 1974-1979
• All CASS patients had coronary angiography at
  enrollment
• Database enrollees treated with CABG or medical
  management based on physician preference
• Identified 3368 patients who required noncardiac
  surgery during 10 years of followup
• Surgeries included general (631), orthopedic
  (214), urologic (327), minor (90)

Eagle, Circulation, 199796
14
CASS

• End points included mortality within 30 days of
  procedure or myocardial infarction
• Among 1961 patients undergoing high risk surgery,
  prior CABG was associated with fewer events when
  compared to medical management
• Death 1.7 versus 3.3 (p0.03)
• MI 0.8 versus 2.7 (p0.002)
• 1297 patients undergoing low risk procedures had
  a mortality of lt1 regardless of prior coronary
  treatment

Eagle, Circulation, 199796
15
Medical Therapy - So What About Beta Blockers?

• Mangano et al (NEJM, 1996)
• Poldermans et al (NEJM, 1999)

16
Medical Therapy

• Mangano et al (NEJM, 1996)
• 200 men at a VA with CAD or at risk for CAD
• Randomized to IV Atenolol 30 minutes prior to
  noncardiac surgery continued throughout
  hsopitalization versus placebo
• No difference in perioperative mortality or
  cardiovascular endpoints
• At 2 years
• Mortality ARR 11, NNT 9
• CV events ARR 15, NNT 7
• 2 year endpoints, aggressive HR control

17
Medical Therapy

• Poldermans et al (NEJM, 1999)
• 112 high-risk patients (positive DSE)
• high-risk major vascular surgeries (AAA repair,
  Aorto-fem bypass)
• Randomized to Bisoprolol 5-10 QD (starting one
  week prior to surgery and continued 30 days
  post-op), or standard care
• Combined 30 day endpoints of mortality or
  non-fatal MI
• ARR 31, NNT 3
• Generalizability, aggressive HR control

18
So Who Gets Beta Blockers?
Fleisher. NEJM, Vol 345, 2001
19
Medical Therapy - Summary

• Very few randomized trials
• Appropriately administered beta blockers may
  reduce the risk of MI and death in high-risk
  patients
• If possible, start beta blockers days or weeks
  before elective surgery
• Titrate dose to a resting HR between 50-60
• Alpha-2 agonists may have similar effects
  (clonidine, mivazerol)

20
(No Transcript)
21
(No Transcript)
22
ACP Guideline

• Does not take into account a patients functional
  status
• Does not tend to recommend stress testing in
  situations when patient is intermediate risk
  where ACC/AHA guideline is more test oriented

23
Mark Wilsons Guideline

• If the patient walked into your clinic (and was
  not going to surgery), would you order a stress
  test based on their history and physical?
• Let sleeping dogs lie.

24
How Well Do We Risk Stratify?

• Prospective Evaluation of Cardiac Risk Indices
  for Patients Undergoing Noncardiac Surgery
  Gilbert et al (Annals, 2000)
• Compared four existing methods for predicting
  perioperative cardiac risk in a prospective
  cohort study
• 2035 patients referred for medical consultation
  before urgent or elective non-cardiac surgery
• Endpoints included cardiac events (MI, angina, or
  acute pulmonary edema), and death (all-cause
  mortality)

25
Gilbert - Methods

• Patients were categorized according to the
  following indices
• American Society of Anesthesiologists (ASA), 1963
• Goldman, 1977
• Detsky (ACP algorithm), 1997
• Canadian Cardiovascular Society (CCS), 1976
• Cardiac events were defined clinically and
  according to generally accepted criteria
• myocardial infarction, unstable angina, acute
  pulmonary edema
• Death defined as all-cause mortality

26
Gilbert - Methods

• To determine the accuracy of the stratification
  systems
• Receiver-operating characteristic (ROC) curves
  were calculated for each index
• Areas under the ROC curve (accuracy) were
  compared

27
Patient Characteristics (2035)
28
Gilbert - Results

• Cardiac Events - 130 (6.4)
• MI - 36 (1.8)
• Pulmonary Edema - 67 (3.3)
• Unstable Angina - 27 (1.3)
• Deaths - 48 (2.4)

29
Gilbert - ResultsAlgorithm Accuracy
30
Gilbert - Conclusions

• None of the models was significantly superior to
  the others
• Cardiac risk indices provide useful clinical
  information about risk, but have limited overall
  accuracy
• There remains room for improvement in our ability
  to determine which patients are at greatest risk
  for cardiac complications

31
But Wait . . .

• Prediction of Perioperative Risk The Glass May
  Be Three-Quarters Full, Eagle et al (Annals,
  2000)
• Tested guidelines do not take into account
  functional status and therefore risk may have
  been underestimated
• Patients may not have had implementation of
  medical therapy based on their risk
• Most importantly, identification of risk is not
  synonymous with prediction of adverse events

32
Eagle - Editorial

• The population studied was one in which there was
  a low prevalence of complications
• Identification of high-risk individuals may have
  led to changes in therapy, and therefore low
  accuracy may be a marker of successful management

33
Lees Revised Cardiac Risk Index

• Lee et al. Derivation and Prospective Validation
  of a Simple Index for Prediction of Cardiac Risk
  of Major Noncardiac Surgery. Circulation, 1999.
• Prospective investigation to derive and validate
  a simple index for the prediction of the risk of
  cardiac complications in major elective
  noncardiac surgery

34
Lee - Methods

• Patients gt 50 years old
• Nonemergent, noncardiac surgery
• All patients had serial CK enzymes and EKGs
  post-operatively
• Major cardiac complications defined as MI,
  pulmonary edema, vfib, cardiac arrest, complete
  heart block

35
Lee - Subjects

• Mean age 66 /- 10 years
• 33 Orthopedic procedures
• 40 High risk procedures
• 16 History of MI

36
Lee - Subjects
37
Lee - Results

• Patients who were using beta blockers at the time
  of admission had similar rates of complications
  as those who were not
• Beta blockers - 2.4 complications
• No beta blockers - 1.8 complications

38
Lee - Results

• 6 independent correlates of major cardiac
  complications were identified in the derivation
  cohort
• Variable Complication Rate
• High risk surgery 3
• Ischemic heart disease 4
• Congestive heart failure 5
• Hx of CVA 6
• Insulin treated DM 6
• Serum creatinine gt 2.0 mg/dl 9

39
Lee - Results
40
Lee - Conclusions

• How the Revised Cardiac Risk Index should be
  used by clinicians remains to be defined
• Possibly perform noninvasive testing on classes
  III and IV

41
Conclusions

• Purpose of preoperative assessment is NOT to
  clear the patient
• Risk stratify with guidelines
• Treat the patient
• High risk patients may benefit from further
  testing
• Remember beta blockers in the appropriate patient
  population